The Economic Impact of Changing the

Texas Medicaid Pharmacy Benefit Structure:

An Analysis of the Potential Effects of
Switching from a Carve-Out to a Carve-In
System or Limiting Network Access

Summary of Key Results from a Study by the Perryman Group

Legislative Advertising: Pharmacy Choice & Access Now, P.O. Box 3435, Arlington, VA 22203, Richard Beck, Board Member
Introduction and Overview

Given tight budget conditions, rising numbers of enrollees in Medicaid, and growing
costs, Texas legislators are considering ways to reduce the cost of Medicaid.
However, any such changes should be carefully analyzed to ensure they do not
lead to future problems and cost escalation.

Texas currently uses a “carve-out” method to provide pharmacy benefits for

Medicaid patients.
o Carve-out systems essentially exclude certain health and pharmacy
services from Medicaid Managed Care.
o The advantages of a carve-out method include the following.
 A carve-out model does not require the presence of layers of
bureaucracy or middlemen (such as Managed Care Organizations
(MCOs) or Prescription Benefit Managers (PBMs)), which helps keep
administrative costs lower. In fact, the current state-run vendor drug
program operates at a low 1% administrative expense level.
 The narrow focus on pharmacy helps optimize costs and service.
Many private-sector firms cite costs and greater choice as benefits of
carving out drug spending, and the majority of large employers carve out
drug benefits. While Medicaid, based on its patient makeup, is very
different from corporate health care, the focus solely on pharmacy-
related management under a carve-out model can also be an advantage
for Medicaid beneficiaries. In fact, the large percentage of cost
associated with blind and disabled recipient’s likely makes it even less
suited to a managed care model than many commercial groups.
 Due to federal law, manufacturers already have to offer their “best
price” for a particular drug for Medicaid fee-for-service programs.
Therefore, prices on drugs under carve-out models are already as low as
other systems.

Recently, proposals have surfaced to change to a “carve-in” mechanism

for funding.
o A carve-in structure involves shifting pharmacy benefits to be included
within the role of the MCOs, which contract with the State to provide health
services for Medicaid programs for a set cost.
o With pharmacy benefits “carved in,” MCOs would likely contract with a PBM, a
for-profit firm which acts as administrator of the prescription drug
program.
o Proponents point to potential savings due to efficiencies under the carve-in
system. However, many of these so-called advantages are illusory.

1
 Moreover, if such an approach is adopted, it is critical that a broad network of
providers be preserved.

Overview of The Perryman Group’s Economic Impact Assessment

From an economic perspective, reductions in fees which threaten the viability of

certain pharmacies (particularly independent establishments and smaller chains
operating in less populous areas) or a switch to a carve in method with
significant network limitations as a basis for providing pharmacy benefits would
lead to significant negative fallout.
o The Perryman Group first measured the effect on pharmacies of various cost-
reduction proposals and found that pharmacy store closings would likely occur.
o These closures would cause economic harm as well as restrict the access of
Medicaid and non-Medicaid patients, leading to further potential negative
effects.
o Three scenarios were formulated to reflect the economic impacts of various
potential cost-reducing actions that have been discussed:
 Scenario I assumes the plan recently put forth by the Texas Health and
Human Services Commission (THHSC) is put in place (along with fee
reductions currently proposed by the Texas Legislature).
 Scenario II reflects the fee cuts discussed by the Texas Legislature (two
1% reductions now in place and an additional $1.00 reduction being
discussed), but assumes that a broad network is maintained irrespective
of whether a “carve-in” or “carve out” approach is implemented.
 Scenario III presumes the implementation of the dispensing fee policy
recommendations embodied in a widely circulated study by The Lewin
Group.

These actions (like any economic activity) would generate multiplier or ripple effects
through the economy. The Perryman Group developed a model some 30 years ago
(with continual updates and refinements since that time) to describe these interactions.
This dynamic input-output assessment model uses a variety of data (from surveys,
industry information, and other sources) to describe the various goods and services
(known as resources or inputs) required to produce another good/service.

In this case, for example, pharmacies regularly purchase products ranging from
supplies to landscaping services. These businesses, in turn, purchase the items
necessary to produce and provide the supplies and services from other companies. In
this way, the pharmacy stores’ effect on the economy ripples out through a variety of
firms across a spectrum of industries. Moreover, the network of pharmacies is crucial

2
 to ensuring adequate access for both Medicaid and Non-Medicaid patients. If access
to needed medications is constrained, additional health care outlays occur, as well as
additional economic disruptions.

Impacts are expressed in terms of several different measures of business activity.

o Total expenditures (or total spending) measures the reduction in dollars
changing hands in the state as a result of the reduced access.
o Gross product (or output) is the reduction in production of goods and services
that will come about in Texas as a result of the changes. This measure is
parallel to the gross domestic product numbers commonly reported by various
media outlets and is a subset of total expenditures.
o Personal income is dollars that end up in the hands of people in the area; the
vast majority of this aggregate derives from the earnings of employees, but
payments such as interest and rents are also included.
o Job losses are expressed as permanent jobs (given that this will be an
ongoing impact).

Foregone economic activity, in turn, results in decreases in fiscal receipts to the State
as well as to local government entities. All results are expressed on an annual basis
as of 2013 in constant (2010) dollars.

Following a discussion of the assumptions and analysis put forth in support of various
plans for changing the structure of Medicaid pharmacy benefits in Texas, a summary
results for key findings from the forthcoming study by The Perryman Group are
presented. The full report includes further information such as disaggregation by
mental and general health, results for Medicaid and Non-Medicaid patients, and
industry-level detail. In addition, further detail regarding the methods used by The
Perryman Group is presented.

Scenario I: THHSC Carve-In Plan with Network Limitations

The recommendation to transfer the Medicaid prescription program to a managed care

(carve-in) system is based on a comparative analysis that requires further evaluation.
The purported net savings is entirely the result of an expected premium tax payment
of $56.1 million over the next biennium. In the absence of that levy, there is a cost
advantage of $11 million from implementing the cost-cutting policies put forth by the
pharmacy industry within the state.

Several assumptions in the analysis by the Texas Health and Human Service
Commission (THHSC) are worthy of additional analysis. Initially, the estimates
suggest an added administrative cost of only $700,000 for the biennium, which is

3
exactly offset by assumed improvements in utilization management. A recent analysis
of the Pennsylvania program found almost $25 million per year in added administrative
expenses, as well as an initial startup cost of $6.7 million for tasks such as notifying
recipients of the changes. The study noted that only about one-third of this amount
could be offset by enhancements in utilization management. While the same
circumstances do not apply in all states, the Pennsylvania system is only about one-
half the size of the Texas program. Moreover, Texas would be maintaining dual
systems in that part of the program would remain fee-for-service. Similarly, even the
percentage increase in administrative costs represented in the Lewin study suggests
added outlay of about $61.6 million per year. These factors combined indicate that the
added fees on a biennium basis in Texas could easily be many times larger than
those used in the comparative analysis. This phenomenon alone could more than
offset the purported savings from the carve-in model.

Similarly, the THHSC comparison suggests that savings over the biennium from a 3%
increase in generic usage would total $11.4 million. Using the cost and number of
subscriptions reported in the Lewin report, the indicated savings is about $48.3 million
per year (almost $100 million per biennium). Even allowing for the federal match and
some lack or comparability, the savings is several times as large as that used in the
comparison. Moreover, with numerous brand name drugs having patents expire in the
near future, the proven success of the current approach in driving generic compliance,
and the incentives available to managed care groups to use brand-name products, it
seems highly likely that this level can be substantially exceeded. While there are other
valid concerns (such as the failure to fully account for the effects of cost-cutting
approaches on premium tax revenues), these major potential discrepancies are
sufficient to illustrate the need for a more thorough investigation before adopting
significant changes.

Perhaps the greatest shortcoming of the comparison is that it does not account for
the dynamic effects of a reduced network that limits access in many areas
(particularly in rural regions), thus leading to complications and added health care
expenses. Some of these implications, which make it imperative that any future
program (whether fee-for-service or managed care) maintain a comprehensive
network, are explored below.

4
The cost-saving measures outlined by the THHSC and the likely associated
network limitations have the potential to cause notable harms to the
pharmaceutical sector including the loss of more than 770 (primarily independent
and small chain) locations, nearly $1.6 billion in annual output (gross product) and
22,135 jobs.

The Annual Impact of the Losses in the Pharmaceutical Sector Associated with
the Potential Network Reductions Accompanying a "Carve-In" Approach to
Medicaid Prescription Management on Business Activity In Texas (as of 2013)

-$2.725 Total Expenditures

-$1.579 Gross Product

-22,135
Permanent -$0.954 Personal Income
Jobs

-$1.168 Retail Sales

-$3.0 -$2.5 -$2.0 -$1.5 -$1.0 -$0.5 $0.0

Billions of 2010 Dollars
Note: Assumes a 30% reduction in network coverage and the dispensing fee reductions currently propsed by the Texas Legislature.
Source: The Perryman Group

5
The total negative effect (including these pharmaceutical losses and incremental
outlays for health care among Medicaid and Non-Medicaid patients) includes $3.1
billion in annual output, 42,923 permanent jobs, and $719.1 million in yearly
dynamic State revenues losses and required outlays. Because of the
disproportionate number of recipients in the blind and disabled category, these effects
could well be even larger.

Looking beyond even the health outcomes, the network limitations of mental health
patients also lead to adverse economic consequences in the form of increased
homelessness, incarceration, and unemployment rates, as well as decreases in
productivity. These phenomena lead to an annual loss of about $3.8 billion in output
and more than 46,500 jobs.

Total Annual Impact on Business Activity in Texas of Pharmaceutical Losses

and Incremental Outlays for Health Care Among All Patients Associated with the
Potential Network Reductions Accompanying a "Carve-In" Approach to
Medicaid Prescription Management

-$5.631 Total Expenditures

-$3.130 Gross Product

-42,923
-$1.974 Personal Income
Permanent
Jobs

-$1.770 Retail Sales

-$7 -$6 -$5 -$4 -$3 -$2 -$1 $0

Billions of 2010 Dollars
Note: Assumes a 30% reduction in network coverage and the dispensing fee reductions currently proposed by the Texas Legislature.
Includes both mental and general health care for both Medicaid and Non-Medicaid patients and assumes that the access effects on
the general population will only be 20% as significant as those for Medicaid recipients.
Source: The Perryman Group

6
Scenario II: Fee Reductions as Proposed in the Texas Legislature

Fee reductions proposed by the Texas Legislature (both the two 1% cuts that
have already occurred and an additional $1.00 reduction) also lead to notable
economic losses, although the consequences are far less severe if the basic
pharmacy network is maintained.

The Perryman Group estimates the reduction in annual business activity within the
pharmaceutical sector to include about 175 closures, $364.4 million in annual
output (gross product) and 5,107 jobs.

The Annual Impact of the Losses in the Pharmaceutical Sector Associated

with the Dispensing Fee Reductions Cirrently Proposed on Business Activity
in Texas (as of 2013)

-$0.629 Total Expenditures

-$0.364 Gross Product

-5,107
Permanent
-$0.220 Personal Income
Jobs

-$0.269 Retail Sales

-$0.7 -$0.6 -$0.5 -$0.4 -$0.3 -$0.2 -$0.1 $0.0

Billions of 2010 Dollars
Source: The Perryman Group

7
 The total negative effect under these assumptions includes $722.3 million in lost
output (gross product) each year, 9,904 permanent jobs, and $165.9 million in
dynamic State fiscal receipts and incremental outlays per annum.

Total Annual Impact on Business Activity in Texas of Pharmaceutical Losses

and Incremental Outlays for Health Care (Mental Health and General) Among All
Patients (Medicaid and Non-Medicaid) Associated with Proposed Dispensing
Fee Reductions

-$1.299 Total Expenditures

-$0.722 Gross Product

-9,904
-$0.456 Personal Income
Permanent
Jobs

-$0.408 Retail Sales

-$1.6 -$1.4 -$1.2 -$1.0 -$0.8 -$0.6 -$0.4 -$0.2 $0.0

Billions of 2010 Dollars

Note: Includes both mental and general health care and assumes that the access effects on the general population will only be 20% as
significant as those for Medicaid recipients.
Source: The Perryman Group

Scenario III: Disbursement Fee Policy Changes as Outlined by The Lewin Group

The Lewin study that has been widely circulated and discussed maintains that Texas
could save approximately $266.3 million annually through switching to a carve-in
system for Medicaid prescription management (including the federal and State
portions). This analysis is flawed in numerous respects.
o First, the majority of the savings is achieved through a dramatic reduction in fees
that would reduce them to less than 20% of the actual cost of dispensing
medications. A decrease of this magnitude would dramatically impact the viability
of the Texas pharmacy sector and have enormous adverse consequences,
essentially dismantling much of the existing pharmacy infrastructure of the state.
In fact, THHSC has previously conducted extensive analysis to demonstrate the
reasonableness of the current fee structure. Moreover, changes in fees can be
(and, in fact, recently have been) implemented within the current structure. The
economic impact of such an approach is outlined below.
o The second major source of purported savings results from an increase in generic
utilization. The Lewin analysis is based on the assumption that Texas has a 69%

8
 use of generics at present and would be able to increase this level through a
"carve-in" program. According to the Texas Health and Human Services
Commission, the current generic use is 72.3%, which is approximately the level
that the Lewin study indicates could be achieved through the "carve-in" system. In
reality, it has already been accomplished with the existing "carve-out" system.
Thus, it is obvious that the purported benefits (1) are overstated and (2) are not
dependent on a "carve-in" model. In fact, independent pharmacies often achieve a
higher level of generic use than Pharmacy Benefit Managers. Because of the
manufacturers’ rebates that PBMs can negotiate (over and above the ones that
flow back to the State and federal governments), they have a specific incentive to
purchase brand-name products.
o Moreover, over 50% of the alleged savings are derived from prescriptions to blind
and disabled recipients. Independent analysis has determined that these
populations are not well suited to managed care pharmacy programs and that such
initiatives are likely to result in larger outlays for medical services.
o Thus, the findings from the Lewin study are not a reliable barometer of the current
environment in Texas and do not form an appropriate mechanism for policy
determination.

9
If the dispensing fee policy as outlined by The Lewin Group is implemented, the
reduction in business activity in the state would be significant. Within the
pharmaceutical sector, such a scenario leads to losses of more than 1150
pharmacies, as well as almost $2.4 billion in annual output and 33,300 jobs.

The Annual Impact of the Losses in the Pharmaceutical Sector Associated with
Implementing the Dispensing Fee Policy Recommendations of the Lewin Study
on Business Activity in Texas (as of 2013)

-$4.103 Total Expenditures

-$2.378 Gross Product

-33,330
Permanent
Jobs -$1.436 Personal Income

-$1.759 Retail Sales

-$4.5 -$4.0 -$3.5 -$3.0 -$2.5 -$2.0 -$1.5 -$1.0 -$0.5 $0.0
Billions of 2010 Dollars
Source: The Perryman Group

10
 Overall losses under this scenario (including pharmaceutical losses and
incremental outlays by patients associated with the potential network
reductions) were estimated to be $4.7 billion in output and 64,632 jobs. Yearly
state revenue losses and expenditure increases were found to total about $1.803
billion.

Total Annual Impact on Business Activity in Texas of Pharmaceutical Losses

and Incremental Outlays for Health Care Associated with Pharmaceutical
Network Effects of Implementing Dispensing Fee Policy Recommendations of
the Lewin Study

-$8.479 Total Expenditures

-$4.714 Gross Product

-64,632
Permanent -$2.973 Personal Income
Jobs

-$2.665 Retail Sales

-$10 -$9 -$8 -$7 -$6 -$5 -$4 -$3 -$2 -$1 $0
Billions of 2010 Dollars
Note: Includes Medicaid and Non-Medicaid patients as well as both mental and general health care. Assumes that the access effec
ts
on the general population will only be 20% as significant as those for Medicaid recipients.
Source: The Perryman Group

Conclusion

Examining potential avenues for savings is a necessary exercise, particularly when

facing a budgetary shortfall. However, maintenance of a vibrant pharmaceutical
network throughout Texas is both cost effective and essential to providing an
adequate health care delivery system.

Proponents of a change to a carve-in system with notable network restrictions purport

that such a switch would benefit the State. However, underlying analyses by The
Lewin Group and the Texas Health and Human Services Commission incorporate
notable flaws or omissions.

Moreover, many aspects of these benefits are achievable and even exceeded by
savings suggested or supported by pharmacy stakeholders. If a carve-in system

11
is implemented, it is imperative that certain provisions be incorporated or excluded
to avoid higher future costs. For example, restricting pharmacy access can lead
to higher costs in terms of care lapses and escalation of medical problems
and expenses.

Changing the Texas system of providing pharmacy benefits to Medicaid

recipients to a carve-in method or otherwise limiting access would involve
substantial negative fallout and should be evaluated within a framework that
fully reflects the overall consequences of policy changes. While the additional
strain on the State budget is dynamic in nature and not always measured in the
framework of policy debates, it is nonetheless very real (as are the associated
costs to the economy and overall health and well-being of Texas citizens).

12
www.rxchoiceandaccess.com www.perrymangroup.com